Mini Project - Sai Teja Burla - Team Vermont :)

Importing Libraries

library("ggpubr")

Loading required package: ggplot2

library("gridExtra")
library("tidyverse")

── Attaching core tidyverse packages ────────────────────────────────────────────────────────────────────────────────────────────────── tidyverse 2.0.0 ──
✔ dplyr     1.1.2     ✔ readr     2.1.4
✔ forcats   1.0.0     ✔ stringr   1.5.0
✔ lubridate 1.9.2     ✔ tibble    3.2.1
✔ purrr     1.0.2     ✔ tidyr     1.3.0── Conflicts ──────────────────────────────────────────────────────────────────────────────────────────────────────────────────── tidyverse_conflicts() ──
✖ dplyr::combine() masks gridExtra::combine()
✖ dplyr::filter()  masks stats::filter()
✖ dplyr::lag()     masks stats::lag()
ℹ Use the ]8;;http://conflicted.r-lib.org/conflicted package]8;; to force all conflicts to become errors

library("ggplot2")
library("ggridges")
library("tidyr")
library("readxl")
library("stringr")
library("ggrepel")
library("gridExtra")
library("patchwork")
library("viridis")

Loading required package: viridisLite

library("broom")

Importing House Price Index Data and Perform some Preprocessing

House.Price.Index <- read_excel("State_and_US_SA.xls", skip = 5)
House.Price.Index <- head(House.Price.Index, - 16) 
House.Price.Index <- 
  data.frame(House.Price.Index %>% 
  dplyr::rename("Date" = "Month") %>% 
  separate_wider_position("Date", c("Year" = 4, "Period" = 3)) %>%
  mutate_at("Year", as.integer) %>%
  mutate_at("AK", as.double)) %>% 
  select(-United.States.seasonally.adjusted)
House.Price.Index

Importing State Region Details

State.Region.Details <- read.table("state_abbrevs.txt", sep = " ", header = TRUE)
State.Region.Details

Importing Consumer Price Index Data

Consumer.Price.Index <- read.csv("cpi.csv", header = TRUE) %>% select(-Series.ID, -Label)
Consumer.Price.Index

Census <- read.csv("combinedcensuspop.csv", header = TRUE) %>%
  select(-GISJOIN, -STATEFP, -STATENH, -NAME) %>%
  filter(YEAR >= 1975 & YEAR <= 2019) %>%
  filter(STATE != 'Puerto Rico') %>%
  filter(STATE != 'District of Columbia') %>%
  dplyr::rename("State" = "STATE", "Population" = "A00AA", "Year" = "YEAR")
Census

Importing Census Data for the Year 1990

Census.1990 <- read.csv("combinedcensuspop.csv", header = TRUE) %>% 
  filter(YEAR == 1990) %>% 
  select(-GISJOIN, -YEAR, -STATEFP, -STATENH, -NAME) %>%
  dplyr::rename("State" = "STATE", "Population" = "A00AA")
Census.1990

Importing Census Data for the Year 2000

Census.2000 <- read.csv("combinedcensuspop.csv", header = TRUE) %>% 
  filter(YEAR == 2000) %>% 
  select(-GISJOIN, -YEAR, -STATEFP, -STATENH, -NAME) %>%
  dplyr::rename("State" = "STATE", "Population" = "A00AA")
Census.2000

Importing Census Data for the Year 2010

Census.2010 <- read.csv("combinedcensuspop.csv", header = TRUE) %>% 
  filter(YEAR == 2010) %>% 
  select(-GISJOIN, -YEAR, -STATEFP, -STATENH, -NAME) %>%
  dplyr::rename("State" = "STATE", "Population" = "A00AA")
Census.2010

Importing Census Data for the Year 2019

Census.2019 <- read.table("censuspop2019.txt", sep = " ", header = TRUE) %>% 
  filter(variable == "POP") %>% 
  select(-GEOID, -variable) %>%
  dplyr::rename("State" = "NAME", "Population" = "value")
Census.2019 <- head(Census.2019, 51)
Census.2019

Importing Density Data for the Year 2019

Density.2019 <- read.table("censuspop2019.txt", sep = " ", header = TRUE) %>% 
  filter(variable == "DENSITY") %>% 
  select(-GEOID, -variable) %>%
  dplyr::rename("State" = "NAME", "Density" = "value") 
Density.2019 <- head(Density.2019, 51)
Density.2019$Density = round(Density.2019$Density, 2)
Density.2019

Combined.2019 <- Census.2019 %>% 
  inner_join(Density.2019, by = join_by(State == State))
Combined.2019

Importing Data for the Year 1975

Combined.1975 <- read.csv("Density1975 - Density1975.csv", header = TRUE) #%>% dplyr::rename("Population.1975" = "Population", "Density.1975" = "Density")
Combined.1975$Density = round((Combined.1975$Density * 100.01) / 99.9, 2)
Combined.1975

Census.1975 <- Combined.1975 %>% select(-Density)
Census.1975

Calculating approximate areas and population densities for intermediate years

Approximate.Areas <- Combined.2019 %>% filter(State != 'District of Columbia')
Approximate.Areas$Area = round(Approximate.Areas$Population / Approximate.Areas$Density, 2)
Approximate.Areas <- Approximate.Areas %>% select(-Population, -Density)
Approximate.Areas

Density.census <- 
  left_join(Census, Approximate.Areas, by = join_by(State == State))
Density.census$Density = Density.census$Population / Density.census$Area
Density.census <- Density.census %>%
  select(-Area)
Density.census

Combined.1975$Year = as.integer(1975)
Combined.2019$Year = as.integer(2019)
Density.census <- rbind(Combined.1975, Density.census, Combined.2019)
Density.census$Density <- round(Density.census$Density, 2)
Density.census <- Density.census %>%
  inner_join(Combined.1975 %>% select(-Year), by = join_by(State == State))
Density.census

Density.census <- Density.census %>%
  inner_join(State.Region.Details, by = join_by(State == State)) %>%
  dplyr::rename("Density" = "Density.x", "Population" = "Population.x")
Density.census

Density.census$Population_Diff = Density.census$Population - Density.census$Population.y
Density.census$Population_Ratio = round(Density.census$Population / Density.census$Population.y, 2)
Density.census$Density_Diff = Density.census$Density - Density.census$Density.y
Density.census$Density_Ratio = round(Density.census$Density / Density.census$Density.y, 2)
Density.census

Density.census <-
  Density.census %>%
  select(-Population.y, -Density.y)
Density.census

Question - 1

houses_long <- data.frame(gather(House.Price.Index, key="State", value="House.Prices", 3:53))
House_Price_Data <- 
  data.frame(houses_long %>%
  inner_join(Consumer.Price.Index, by = c("Year", "Period")) %>%
  dplyr::rename("CPI.Values" = "Value"))
House_Price_Data

compare_2019_1975 <- 
  data.frame(House_Price_Data %>%
  left_join(State.Region.Details, by = join_by(State == Code)) %>%
  dplyr::rename("Code" = "State", "State" = "State.y"))
compare_2019_1975$Adjusted.House.Prices = (compare_2019_1975$House.Prices/compare_2019_1975$CPI.Values) * 100
compare_2019_1975$State <- replace(compare_2019_1975$State, is.na(compare_2019_1975$State), "US Adjusted Avg.")
compare_2019_1975$Region <- replace(compare_2019_1975$Region, is.na(compare_2019_1975$Region), "US Adjusted Avg.")
compare_2019_1975

ggplot(data = compare_2019_1975) +
  geom_point(aes(x = Period, y = Adjusted.House.Prices, color=State), size = 0.3, alpha=0.3) +
  facet_grid(~ Year, shrink = TRUE) +
  scale_color_discrete(guide = "none") +
  theme_classic() +
  ggtitle("House Prices (1975 - 2019) for All States") +
  xlab("Time Period (Years)") + ylab("Adjusted House Prices") +
  labs(subtitle = "General trends in house prices (adjusted for inflation) for all states") +
  theme(strip.text = element_text(angle = 90, hjust = 0.5, size = 7), axis.text.x = element_blank(), panel.spacing = unit(0.0, "lines")) 

grouped_year <-
  data.frame(compare_2019_1975 %>%
  dplyr::group_by(Year, Code, Region, State) %>%
  dplyr::summarise(Average.House.Prices = mean(Adjusted.House.Prices), .groups = "keep"))
grouped_year

Density.census <- Density.census %>%
  inner_join(grouped_year %>% select(-Code, -Region), by = join_by(Year == Year, State == State))
Density.census

P.1975 <- Density.census %>%
  filter(Year == 1975) %>%
  select(State, Average.House.Prices) %>%
  dplyr::rename("hp1975" = "Average.House.Prices")
Density.census <- Density.census %>%
  left_join(P.1975, by = join_by(State == State))
Density.census

Density.census$Price_Diff = Density.census$Average.House.Prices - Density.census$hp1975
Density.census$Price_Ratio = Density.census$Average.House.Prices / Density.census$hp1975
Density.census <- Density.census %>% select(-hp1975)
Density.census

ggplot(data = grouped_year %>% filter(State != 'District of Columbia')) +
  geom_smooth(method = 'lm', formula = y~x, se = FALSE, color = 'darkgrey', linewidth = 0.4, aes(x = Year, y = Average.House.Prices)) +
  geom_line(stat = 'identity', aes(x = Year, y = Average.House.Prices, color = Region)) +
  facet_wrap(~State, ncol = 10) +
  theme_classic() +
  ggtitle("House Prices (1975 - 2019) for All States") +
  xlab("Time Period (Years)") + ylab("Adjusted Average House Prices") +
  labs(subtitle = "Trends in house prices (adjusted for inflation) faceted by state") +
  theme(strip.text = element_text(size = 8), axis.text.x = element_blank(), axis.text.y = element_blank(), panel.spacing = unit(0.1, "lines"))

temp <- grouped_year %>%
  filter(Year == 1975) %>%
  select(State, Average.House.Prices) %>%
  dplyr::rename('1975.Prices' = 'Average.House.Prices')
gy <- left_join(grouped_year, temp, by = join_by(State == State))
gy

ggplot(data = gy %>% filter(State != "District of Columbia"), aes(x = Year, y = Average.House.Prices - `1975.Prices`)) +
  geom_smooth(method = 'lm', formula = y~x, se = FALSE, color = 'darkgrey', linetype = "dashed", linewidth = 1) +
  geom_line(stat = 'identity', aes(color = State)) +
  geom_hline(yintercept = 0, color = 'black') +
  facet_wrap(~Region, ncol = 1) +
  scale_color_discrete(guide = "none") +
  theme_classic() +
  ggtitle("House Prices (1975 - 2019) for All States") +
  xlab("Time Period (Years)") + ylab("Difference in House Prices") +
  labs(subtitle = "Differences in house prices (adjusted for inflation) from 1975 for all states faceted by region") +
  theme(strip.text = element_text(size = 6), axis.text.x = element_blank(), panel.spacing = unit(0.1, "lines"))

ggplot(data = grouped_year %>% filter(State != "District of Columbia"), aes(x = Year, y = Average.House.Prices)) +
  geom_smooth(method = 'lm', formula = y~x, se = FALSE, color = 'darkgrey', linetype = "dashed", linewidth = 1) +
  geom_line(stat = 'identity', aes(color = State)) +
  facet_wrap(~Region, ncol = 1) +
  scale_color_discrete(guide = "none") +
  theme_classic() +
  ggtitle("House Prices (1975 - 2019) for All States") +
  xlab("Time Period (Years)") + ylab("Adjusted Average House Prices") +
  labs(subtitle = "Trends in house prices (adjusted for inflation) for all states faceted by region") +
  theme(strip.text = element_text(size = 6), axis.text.x = element_blank(), panel.spacing = unit(0.1, "lines"))

grouped_region <-
  data.frame(compare_2019_1975 %>% filter(State != "District of Columbia") %>%
  dplyr::group_by(Year, Region) %>%
  dplyr::summarise(Average.House.Prices = mean(Adjusted.House.Prices), .groups = "keep"))
grouped_region

ggplot(data = grouped_region, aes(x = Year, y = Average.House.Prices, color=Region)) +
  geom_smooth(method = 'lm', formula = y~x, se = FALSE, color = 'darkgrey', linetype = "dashed", linewidth = 1) +
  geom_line(stat = "identity", alpha=0.6) +
  theme_classic() +
  ggtitle("House Prices (1975 - 2019) for All Regions") +
  xlab("Time Period (Years)") + ylab("Adjusted Average House Prices") +
  labs(subtitle = "Trends in average house prices (adjusted for inflation) grouped by region") +
  theme(strip.text = element_text(angle = 45, hjust = 0.5, size = 6), panel.spacing = unit(0.0, "lines"))

grouped_region_t <-
  data.frame(gy %>% filter(State != "District of Columbia") %>%
  dplyr::group_by(Year, Region) %>%
  dplyr::summarise(Average.House.Prices = mean(Average.House.Prices), Prices.1975 = mean(`1975.Prices`), .groups = "keep"))
grouped_region_t

ggplot(data = grouped_region_t, aes(x = Year, y = Average.House.Prices - Prices.1975, color=Region)) +
  geom_smooth(method = 'lm', formula = y~x, se = FALSE, color = 'darkgrey', linetype = "dashed", linewidth = 1) +
  geom_hline(yintercept = 0, color = 'black') +
  geom_line(stat = "identity", alpha=0.6) +
  theme_classic() +
  ggtitle("House Prices (1975 - 2019) for All Regions") +
  xlab("Time Period (Years)") + ylab("Difference in House Prices") +
  labs(subtitle = "DIference in average house prices from 1975 to 2019 grouped by region") +
  theme(strip.text = element_text(angle = 45, hjust = 0.5, size = 6), panel.spacing = unit(0.0, "lines"))

grouped_region_wide <- 
  grouped_region %>%
  pivot_wider(names_from = Year, values_from = Average.House.Prices)
grouped_region_wide$Price.Difference = grouped_region_wide$'2019' - grouped_region_wide$'1975'
grouped_region_wide <- 
  grouped_region_wide %>%
  arrange(Price.Difference)
grouped_region_wide

ggplot(data = grouped_region_wide %>% filter(Region != "US Adjusted Avg."), aes(x = Region, y = Price.Difference, fill=Region)) +
  geom_bar(stat = "identity", alpha=0.6) +
  theme_classic() +
  ggtitle("House Prices (1975 - 2019) for All Regions") +
  xlab("Region") + ylab("Price Difference") +
  labs(subtitle = "Difference in average house prices for each region from 1975 to 2019")

grouped_year_wide <- 
  grouped_year %>%
  pivot_wider(names_from = Year, values_from = Average.House.Prices)
grouped_year_wide$Price.Difference = grouped_year_wide$'2019' - grouped_year_wide$'1975'
grouped_year_wide <- 
  grouped_year_wide %>%
  arrange(Price.Difference) %>%
  filter(State != 'District of Columbia')
grouped_year_wide

most_price_increase <- tail(grouped_year_wide, 10)
most_price_drop <- head(grouped_year_wide, 10)
most_price_increase$var = "Price Increase"
most_price_drop$var = "Price Decrease"
price_change = rbind(head(most_price_drop, 5), tail(most_price_increase, 5))
price_change

bp1 <- ggplot(data = head(most_price_drop, 5), aes(x = State, y = Price.Difference)) +
  geom_bar(stat = "identity", fill = "darkblue", alpha = 0.7) +
  geom_label_repel(aes(label = round(Price.Difference, 2)), nudge_x = -0.1, size = 3) +
  theme_classic() +
  ggtitle("Least Price Change") +
  xlab("State") + ylab("Price Difference") +
  labs(subtitle = "States with the lowest (most negative) price change between 1975 and 2019")
bp2 <- ggplot(data = tail(most_price_increase, 5), aes(x = State, y = Price.Difference)) +
  geom_bar(stat = "identity", fill = "darkgreen", alpha = 0.7) +
  geom_label_repel(aes(label = round(Price.Difference, 2)), nudge_x = 0.1, size = 3) +
  theme_classic() +
  ggtitle("Most Price Change") +
  xlab("State") + ylab("Price Difference") +
  labs(subtitle = "States with the highest (most positive) price change between 1975 and 2019")
grid.arrange(bp1, bp2, ncol = 2)

Question - 2

Combined.2019 <- Census.2019 %>% 
  inner_join(Density.2019, by = join_by(State == State)) %>% 
  inner_join(grouped_year %>% filter(Year == 2019) %>% select(State, Average.House.Prices), by = join_by(State == State))
Combined.2019$Year = 2019
Combined.2019 <- Combined.2019 %>%
  filter(State != 'District of Columbia')
Combined.2019

Combined <-
  Combined.2019 %>%
  dplyr::rename("Population.2019" = "Population", "Density.2019" = "Density", "Prices.2019" = "Average.House.Prices") %>%
  inner_join(Combined.1975, by = join_by(State == State)) %>%
  inner_join(grouped_year %>% filter(Year == 1975) %>% select(State, Average.House.Prices), by = join_by(State == State)) %>%
  dplyr::rename("Population.1975" = "Population", "Density.1975" = "Density", "Prices.1975" = "Average.House.Prices") %>%
  select(-Year.x, -Year.y) %>%
  inner_join(State.Region.Details, by = join_by(State == State))
Combined$Price.Diff = Combined$Prices.2019 - Combined$Prices.1975
Combined$Density.Diff = Combined$Density.2019 - Combined$Density.1975
Combined$Population.Diff = Combined$Population.2019 - Combined$Population.1975
Combined

a <- grouped_year %>%
  filter(Year == 1975) %>%
  select(Year, State, Average.House.Prices) %>%
  inner_join(Combined.1975 %>% select(-Year), by = join_by(State == State))
b <-
  Combined.2019 %>%
  select(Year, State, Average.House.Prices, Population, Density)
Combined_long <- rbind(a, b) %>% 
  filter(State != "District of Columbia") %>% 
  dplyr::rename("Price" = "Average.House.Prices") %>% 
  mutate_at("Year", as.character) %>%
  inner_join(State.Region.Details, by = join_by(State == State))
Combined_long

ggplot(data = Combined, aes(x = Density.Diff, y = Price.Diff, size = Population.Diff, color = State)) +
  geom_smooth(method = 'lm', formula = y~x, color = 'darkgrey', alpha = 0.4, linetype = 'dashed', linewidth = 1, se = FALSE) +
  geom_point(alpha = 0.5) + 
  scale_size_continuous(guide = "none") +
  scale_color_discrete(guide = "none") +
  geom_label_repel(aes(label = Code), nudge_x = 0.1, nudge_y = 0.1, size = 2, max.overlaps = 20) +
  theme_classic() +
  ggtitle("Trend Between Density and House Prices") +
  xlab("Difference in Density") + ylab("Price Difference") +
  labs(subtitle = "Scatterplot explaining the relationship between the difference in densities and the change in house prices between 1975 and 2019")

ggplot(data = Combined, aes(x = Density.Diff, y = Price.Diff, size = Population.Diff, color = State)) +
  geom_smooth(method = 'lm', formula = y~x, color = 'darkgrey', alpha = 0.4, linetype = 'dashed', linewidth = 1, se = FALSE) +
  geom_point(alpha = 0.5) + 
  scale_size_continuous(guide = "none") +
  scale_color_discrete(guide = "none") +
  geom_label_repel(aes(label = Code), nudge_x = 0.1, nudge_y = 0.1, size = 2, max.overlaps = 20) +
  facet_wrap(~Region, nrow = 2) +
  theme_classic() +
  ggtitle("Trend Between Density and House Prices") +
  xlab("Difference in Density") + ylab("Price Difference") +
  labs(subtitle = "Scatterplot explaining the relationship between the difference in densities and the change in house prices between 1975 and 2019 for each region") +
  theme(strip.text = element_text(size = 8), panel.spacing = unit(1, "lines"))

ggplot(data = Combined, aes(x = log((Density.2019/Density.1975)), y = log((Prices.2019/Prices.1975)), size = (Population.2019/Population.1975), color = State)) +
  geom_hline(yintercept = 0, color = 'black', linetype = 'dotted', linewidth = 0.4) +
  geom_vline(xintercept = 0, color = 'black', linetype = 'dotted', linewidth = 0.4) +
  geom_smooth(method = 'lm', formula = y~x, color = 'darkgrey', alpha = 0.4, linetype = 'dashed', linewidth = 1, se = FALSE) +
  geom_point(alpha = 0.5) + 
  scale_size_continuous(guide = "none") +
  scale_color_discrete(guide = "none") +
  geom_label_repel(aes(label = Code), nudge_x = 0.01, nudge_y = 0.01, size = 2, max.overlaps = 20) +
  theme_classic() +
  ggtitle("Trend Between Density and House Prices") +
  xlab("Proportion of Density (log scale)") + ylab("Proportion of Price (log scale)") +
  labs(subtitle = "Scatterplot explaining the relationship between the proportional change in densities and the proportional change in house prices between 1975 and 2019")

ggplot(data = Combined, aes(x = log((Density.2019/Density.1975)), y = log((Prices.2019/Prices.1975)), size = (Population.2019/Population.1975), color = State)) +
  geom_hline(yintercept = 0, color = 'black', linetype = 'dotted', linewidth = 0.4) +
  geom_vline(xintercept = 0, color = 'black', linetype = 'dotted', linewidth = 0.4) +
  geom_smooth(method = 'lm', formula = y~x, color = 'darkgrey', alpha = 0.4, linetype = 'dashed', linewidth = 1, se = FALSE) +
  geom_point(alpha = 0.5) + 
  scale_size_continuous(guide = "none") +
  scale_color_discrete(guide = "none") +
  geom_label_repel(aes(label = Code), nudge_x = 0.1, nudge_y = 0.1, size = 2, max.overlaps = 20) +
  facet_wrap(~Region, nrow = 2) +
  theme_classic() +
  ggtitle("Trend Between Density and House Prices") +
  xlab("Proportion of Density (log scale)") + ylab("Proportion of Price (log scale)") +
  labs(subtitle = "Scatterplot explaining the relationship between the proportional change in densities and the proportional change in house prices between 1975 and 2019 faceted by region") +
  theme(strip.text = element_text(size = 8), panel.spacing = unit(1, "lines"))

ggplot(data = Density.census %>% filter(Year != 1975), aes(x = log(Density_Ratio), y = log(Price_Ratio), color = Code, )) +
  geom_point(alpha = 0.4) + 
  geom_hline(yintercept = 0, color = 'black', linetype = 'dotted', linewidth = 0.4) +
  geom_vline(xintercept = 0, color = 'black', linetype = 'dotted', linewidth = 0.4) +
  geom_smooth(method = 'lm', color = 'black', se = FALSE, linewidth = 0.6) +
  #geom_label_repel(aes(label = Code, color = Code), size = 2, max.overlaps = 50) +
  scale_color_discrete(guide = "none") +
  facet_grid(Region~Year) + 
  theme_bw() +
  ggtitle("Trend Between Density and House Prices") +
  xlab("Proportion of Density (log scale)") + ylab("Proportion of Price (log scale)") +
  labs(subtitle = "Proportional change in densities and the proportional change in house prices between 1975 and 2019 faceted by region and year") +
  theme(panel.grid = element_blank())

ggplot(data = Density.census, aes(x = Density, y = Average.House.Prices, color = Region)) +
  geom_point(alpha = 0.4) + 
  #geom_hline(yintercept = 1, color = 'darkgray', linetype = 'dashed', linewidth = 0.4) +
  #geom_vline(xintercept = 1, color = 'darkgray', linetype = 'dashed', linewidth = 0.4) +
  geom_smooth(method = 'lm', color = 'black', se = FALSE, linewidth = 0.6) +
  #geom_label_repel(aes(label = Code, color = Code), size = 2, max.overlaps = 50) +
  #scale_color_discrete(guide = "none") +
  facet_grid(~Year) + 
  theme_bw() +
  ggtitle("Trend Between Density and House Prices") +
  xlab("Population Density") + ylab("Average House Prices") +
  labs(subtitle = "Proportional change in densities and the proportional change in house prices across decades between 1975 and 2019 faceted by region") +
  theme(panel.grid = element_blank())

ggplot(data = Density.census, aes(x = Year, y = log(Price_Ratio), color = log(Density_Ratio))) +
  geom_line(linewidth = 1) +
  scale_color_distiller(name = 'Proportion of Density', palette = "YlOrRd") +
  facet_wrap(~State, nrow = 5) + 
  theme_bw() +
  ggtitle("Trend Between Density and House Prices") +
  xlab("Years") + ylab("Proportion of Price (log scale)") +
  labs(subtitle = "Scatterplot explaining the relationship between the proportional change in densities and the proportional change in house prices between 1975 and 2019 faceted by region") +
  theme(panel.grid = element_blank(), axis.text.x = element_text(hjust = 1, angle = 45))

Question - 3

Year Range - 1990 to 2000

a1 <- 
  rbind(
    (grouped_year %>%
    filter(Year == 1990) %>%
    inner_join(Census.1990, by = join_by(State == State))),
    (grouped_year %>%
    filter(Year == 2000) %>%
    inner_join(Census.2000, by = join_by(State == State)))) %>%
    select(-Average.House.Prices) %>%
    pivot_wider(names_from = Year, values_from = Population) %>%
    dplyr::rename("Population.1990" = "1990", "Population.2000" = "2000")
b1 <-
  rbind(
    (grouped_year %>%
    filter(Year == 1990) %>%
    inner_join(Census.1990, by = join_by(State == State))),
    (grouped_year %>%
    filter(Year == 2000) %>%
    inner_join(Census.2000, by = join_by(State == State)))) %>%
    select(-Population) %>%
    pivot_wider(names_from = Year, values_from = Average.House.Prices) %>%
    dplyr::rename("Average.House.Price.1990" = "1990", "Average.House.Price.2000" = "2000")
Values.1990.2000 <-
  a1 %>% 
  inner_join(b1, by = join_by(State)) %>% 
  select(-Code.y, -Region.y) %>% 
  dplyr::rename("Code" = "Code.x", "Region" = "Region.x") %>%
  filter(State != "District of Columbia")
Values.1990.2000

Values.1990.2000 <- 
  Values.1990.2000 %>%
  arrange(Average.House.Price.2000 - Average.House.Price.1990)

bp3 <- ggplot(data = head(Values.1990.2000, 5), aes(x = State, y = ((Average.House.Price.2000 - Average.House.Price.1990)/Average.House.Price.1990)*100)) +
  geom_bar(stat = "identity", fill = "darkblue", alpha = 0.7) +
  geom_label_repel(aes(label = round(((Average.House.Price.2000 - Average.House.Price.1990)/Average.House.Price.1990)*100, 2)), nudge_x = -0.1, size = 3) +
  theme_classic() +
  ggtitle("Least Price Change") +
  xlab("State") + ylab("Price Difference %") +
  labs(subtitle = "States with the lowest percentage price change between 1990 and 2000")
bp4 <- ggplot(data = tail(Values.1990.2000, 5), aes(x = State, y = ((Average.House.Price.2000 - Average.House.Price.1990)/Average.House.Price.1990)*100)) +
  geom_bar(stat = "identity", fill = "darkgreen", alpha = 0.7) +
  geom_label_repel(aes(label = round(((Average.House.Price.2000 - Average.House.Price.1990)/Average.House.Price.1990)*100, 2)), nudge_x = 0.1, size = 3) +
  theme_classic() +
  ggtitle("Most Price Change") +
  xlab("State") + ylab("Price Difference %") +
  labs(subtitle = "States with the highest percentage price change between 1990 and 2000")
grid.arrange(bp3, bp4, ncol = 2)

Values.1990.2000 <- 
  Values.1990.2000 %>%
  arrange(Population.2000 - Population.1990)

bp5 <- ggplot(data = head(Values.1990.2000, 5), aes(x = State, y = ((Population.2000 - Population.1990)/Population.1990)*100)) +
  geom_bar(stat = "identity", fill = "darkblue", alpha = 0.7) +
  geom_label_repel(aes(label = round(((Population.2000 - Population.1990)/Population.1990)*100, 2)), nudge_x = -0.1, size = 3) +
  theme_classic() +
  ggtitle("Least Population Change") +
  xlab("State") + ylab("Population Difference %") +
  labs(subtitle = "States with the lowest percentage population change between 1990 and 2000")
bp6 <- ggplot(data = tail(Values.1990.2000, 5), aes(x = State, y = ((Population.2000 - Population.1990)/Population.1990)*100)) +
  geom_bar(stat = "identity", fill = "darkgreen", alpha = 0.7) +
  geom_label_repel(aes(label = round(((Population.2000 - Population.1990)/Population.1990)*100, 2)), nudge_x = 0.1, size = 3) +
  theme_classic() +
  ggtitle("Most Population Change") +
  xlab("State") + ylab("Population Difference %") +
  labs(subtitle = "States with the highest percentage population change between 1990 and 2000")
grid.arrange(bp5, bp6, ncol = 2)

ggplot(data = Values.1990.2000, aes(x = (Population.2000 - Population.1990), y = (Average.House.Price.2000 - Average.House.Price.1990), color = State)) +
  geom_smooth(method = 'lm', formula = y~x, color = 'darkgrey', alpha = 0.4, linetype = 'dashed', linewidth = 1, se = FALSE) +
  geom_point(alpha = 0.5) + 
  scale_size_continuous(guide = "none") +
  scale_color_discrete(guide = "none") +
  geom_label_repel(aes(label = Code), size = 2, max.overlaps = 20) +
  theme_classic() +
  ggtitle("(1990 and 2000) Trend Between Population and House Prices") +
  xlab("Difference in Price") + ylab("Difference in House Prices") +
  labs(subtitle = "Scatterplot explaining the relationship between the difference in population and the difference in house prices between 1990 and 2000")

ggplot(data = Values.1990.2000, aes(x = (Population.2000 - Population.1990), y = (Average.House.Price.2000 - Average.House.Price.1990), color = State)) +
  geom_smooth(method = 'lm', formula = y~x, color = 'darkgrey', alpha = 0.4, linetype = 'dashed', linewidth = 1, se = FALSE) +
  geom_point(alpha = 0.5) + 
  scale_size_continuous(guide = "none") +
  scale_color_discrete(guide = "none") +
  geom_label_repel(aes(label = Code), size = 2, max.overlaps = 20) +
  facet_wrap(~Region, nrow = 2) +
  theme_classic() +
  ggtitle("(1990 and 2000) Trend Between Population and House Prices") +
  xlab("Difference in Price") + ylab("Difference in House Prices") +
  labs(subtitle = "Scatterplot explaining the relationship between the difference in population and the difference in house prices between 1990 and 2000 faceted by region") +
  theme(strip.text = element_text(size = 8), panel.spacing = unit(1, "lines"))

ggplot(data = Values.1990.2000, aes(x = log((Population.2000 / Population.1990)), y = log((Average.House.Price.2000 / Average.House.Price.1990)), color = State)) +
  geom_hline(yintercept = 0, color = 'black', linetype = 'dotted', linewidth = 0.4) +
  geom_vline(xintercept = 0, color = 'black', linetype = 'dotted', linewidth = 0.4) +
  geom_smooth(method = 'lm', formula = y~x, color = 'darkgrey', alpha = 0.4, linetype = 'dashed', linewidth = 1, se = FALSE) +
  geom_point(alpha = 0.5) + 
  scale_size_continuous(guide = "none") +
  scale_color_discrete(guide = "none") +
  geom_label_repel(aes(label = Code), size = 2, max.overlaps = 20) +
  theme_classic() +
  ggtitle("(1990 and 2000) Trend Between Population and House Prices") +
  xlab("Proportion of Population (log scale)") + ylab("Proportion of House Prices (log scale)") +
  labs(subtitle = "Scatterplot explaining the relationship between the ratio of population and the ratio of house prices between 1990 and 2000")

ggplot(data = Values.1990.2000, aes(x = log((Population.2000 / Population.1990)), y = log((Average.House.Price.2000 / Average.House.Price.1990)), color = State)) +
  geom_hline(yintercept = 0, color = 'black', linetype = 'dotted', linewidth = 0.4) +
  geom_vline(xintercept = 0, color = 'black', linetype = 'dotted', linewidth = 0.4) +
  geom_smooth(method = 'lm', formula = y~x, color = 'darkgrey', alpha = 0.4, linetype = 'dashed', linewidth = 1, se = FALSE) +
  geom_point(alpha = 0.5) + 
  scale_size_continuous(guide = "none") +
  scale_color_discrete(guide = "none") +
  geom_label_repel(aes(label = Code), size = 2, max.overlaps = 50) +
  facet_wrap(~Region, nrow = 2) +
  theme_classic() +
  ggtitle("(1990 and 2000) Trend Between Population and House Prices") +
  xlab("Proportion of Population (log scale)") + ylab("Proportion of House Prices (log scale)") +
  labs(subtitle = "Scatterplot explaining the relationship between the ratio of population and the ratio of house prices between 1990 and 2000 faceted by region") +
  theme(strip.text = element_text(size = 8), panel.spacing = unit(1, "lines"))

Year Range - 2000 to 2010

a2 <- 
  rbind(
    (grouped_year %>%
    filter(Year == 2000) %>%
    inner_join(Census.2000, by = join_by(State == State))),
    (grouped_year %>%
    filter(Year == 2010) %>%
    inner_join(Census.2010, by = join_by(State == State)))) %>%
    select(-Average.House.Prices) %>%
    pivot_wider(names_from = Year, values_from = Population) %>%
    dplyr::rename("Population.2000" = "2000", "Population.2010" = "2010")
b2 <-
  rbind(
    (grouped_year %>%
    filter(Year == 2000) %>%
    inner_join(Census.2010, by = join_by(State == State))),
    (grouped_year %>%
    filter(Year == 2010) %>%
    inner_join(Census.2010, by = join_by(State == State)))) %>%
    select(-Population) %>%
    pivot_wider(names_from = Year, values_from = Average.House.Prices) %>%
    dplyr::rename("Average.House.Price.2000" = "2000", "Average.House.Price.2010" = "2010")
Values.2000.2010 <-
  a2 %>% 
  inner_join(b2, by = join_by(State)) %>% 
  select(-Code.y, -Region.y) %>%
  dplyr::rename("Code" = "Code.x", "Region" = "Region.x") %>%
  filter(State != "District of Columbia")
Values.2000.2010

Values.2000.2010 <- 
  Values.2000.2010 %>%
  arrange(Average.House.Price.2010 - Average.House.Price.2000)

bp7 <- ggplot(data = head(Values.2000.2010, 5), aes(x = State, y = ((Average.House.Price.2010 - Average.House.Price.2000)/Average.House.Price.2000)*100)) +
  geom_bar(stat = "identity", fill = "darkblue", alpha = 0.7) +
  geom_label_repel(aes(label = round(((Average.House.Price.2010 - Average.House.Price.2000)/Average.House.Price.2000)*100, 2)), nudge_x = -0.1, size = 3) +
  theme_classic() +
  ggtitle("Least Price Change") +
  xlab("State") + ylab("Price Difference %") +
  labs(subtitle = "States with the lowest percentage price change between 2000 and 2010")
bp8 <- ggplot(data = tail(Values.2000.2010, 5), aes(x = State, y = ((Average.House.Price.2010 - Average.House.Price.2000)/Average.House.Price.2000)*100)) +
  geom_bar(stat = "identity", fill = "darkgreen", alpha = 0.7) +
  geom_label_repel(aes(label = round(((Average.House.Price.2010 - Average.House.Price.2000)/Average.House.Price.2000)*100, 2)), nudge_x = 0.1, size = 3) +
  theme_classic() +
  ggtitle("Most Price Change") +
  xlab("State") + ylab("Price Difference %") +
  labs(subtitle = "States with the highest percentage price change between 2000 and 2010")
grid.arrange(bp7, bp8, ncol = 2)

Values.2000.2010 <- 
  Values.2000.2010 %>%
  arrange(Population.2010 - Population.2000)

bp9 <- ggplot(data = head(Values.2000.2010, 5), aes(x = State, y = ((Population.2010 - Population.2000)/Population.2000)*100)) +
  geom_bar(stat = "identity", fill = "darkblue", alpha = 0.7) +
  geom_label_repel(aes(label = round(((Population.2010 - Population.2000)/Population.2000)*100, 2)), nudge_x = -0.1, size = 3) +
  theme_classic() +
  ggtitle("Least Population Change") +
  xlab("State") + ylab("Price Difference %") +
  labs(subtitle = "States with the lowest percentage population change between 2000 and 2010")
bp10 <- ggplot(data = tail(Values.2000.2010, 5), aes(x = State, y = ((Population.2010 - Population.2000)/Population.2000)*100)) +
  geom_bar(stat = "identity", fill = "darkgreen", alpha = 0.7) +
  geom_label_repel(aes(label = round(((Population.2010 - Population.2000)/Population.2000)*100, 2)), nudge_x = 0.1, size = 3) +
  theme_classic() +
  ggtitle("Most Population Change") +
  xlab("State") + ylab("Price Difference %") +
  labs(subtitle = "States with the highest percentage population change between 2000 and 2010")
grid.arrange(bp9, bp10, ncol = 2)

ggplot(data = Values.2000.2010, aes(x = (Population.2010 - Population.2000), y = (Average.House.Price.2010 - Average.House.Price.2000), color = State)) +
  geom_smooth(method = 'lm', formula = y~x, color = 'darkgrey', alpha = 0.4, linetype = 'dashed', linewidth = 1, se = FALSE) +
  geom_point(alpha = 0.5) + 
  scale_size_continuous(guide = "none") +
  scale_color_discrete(guide = "none") +
  geom_label_repel(aes(label = Code), size = 2, max.overlaps = 20) +
  theme_classic() +
  ggtitle("(2000 and 2010) Trend Between Population and House Prices") +
  xlab("Difference in Population") + ylab("Difference in House Prices") +
  labs(subtitle = "Scatterplot explaining the relationship between the difference in population and the difference in house prices between 2000 and 2010")

ggplot(data = Values.2000.2010, aes(x = (Population.2010 - Population.2000), y = (Average.House.Price.2010 - Average.House.Price.2000), color = State)) +
  geom_smooth(method = 'lm', formula = y~x, color = 'darkgrey', alpha = 0.4, linetype = 'dashed', linewidth = 1, se = FALSE) +
  geom_point(alpha = 0.5) + 
  scale_size_continuous(guide = "none") +
  scale_color_discrete(guide = "none") +
  geom_label_repel(aes(label = Code), size = 2, max.overlaps = 20) +
  facet_wrap(~Region, nrow = 2) +
  theme_classic() +
  ggtitle("(2000 and 2010) Trend Between Population and House Prices") +
  xlab("Difference in Population") + ylab("Difference in House Prices") +
  labs(subtitle = "Scatterplot explaining the relationship between the difference in population and the difference in house prices between 2000 and 2010 faceted by region") +
  theme(strip.text = element_text(size = 8), panel.spacing = unit(1, "lines"))

ggplot(data = Values.2000.2010, aes(x = log((Population.2010 / Population.2000)), y = log((Average.House.Price.2010 / Average.House.Price.2000)), color = State)) +
  geom_hline(yintercept = 0, color = 'black', linetype = 'dotted', linewidth = 0.4) +
  geom_vline(xintercept = 0, color = 'black', linetype = 'dotted', linewidth = 0.4) +
  geom_smooth(method = 'lm', formula = y~x, color = 'darkgrey', alpha = 0.4, linetype = 'dashed', linewidth = 1, se = FALSE) +
  geom_point(alpha = 0.5) + 
  scale_size_continuous(guide = "none") +
  scale_color_discrete(guide = "none") +
  geom_label_repel(aes(label = Code), size = 2, max.overlaps = 20) +
  theme_classic() +
  ggtitle("(2000 and 2010) Trend Between Population and House Prices") +
  xlab("Proportion of Population (log scale)") + ylab("Proportion of House Prices (log scale)") +
  labs(subtitle = "Scatterplot explaining the relationship between the ratio of population and the ratio of house prices between 2000 and 2010")

ggplot(data = Values.2000.2010, aes(x = log((Population.2010 / Population.2000)), y = log((Average.House.Price.2010 / Average.House.Price.2000)), color = State)) +
  geom_hline(yintercept = 0, color = 'black', linetype = 'dotted', linewidth = 0.4) +
  geom_vline(xintercept = 0, color = 'black', linetype = 'dotted', linewidth = 0.4) +
  geom_smooth(method = 'lm', formula = y~x, color = 'darkgrey', alpha = 0.4, linetype = 'dashed', linewidth = 1, se = FALSE) +
  geom_point(alpha = 0.5) + 
  scale_size_continuous(guide = "none") +
  scale_color_discrete(guide = "none") +
  geom_label_repel(aes(label = Code), size = 2, max.overlaps = 20) +
  facet_wrap(~Region, nrow = 2) +
  theme_classic() +
  ggtitle("(2000 and 2010) Trend Between Population and House Prices") +
  xlab("Proportion of Population (log scale)") + ylab("Proportion of House Prices (log scale)") +
  labs(subtitle = "Scatterplot explaining the relationship between the ratio of population and the ratio of house prices between 2000 and 2010 faceted by region") +
  theme(strip.text = element_text(size = 8), panel.spacing = unit(1, "lines"))

Year Range - 2010 to 2019

a3 <- 
  rbind(
    (grouped_year %>%
    filter(Year == 2010) %>%
    inner_join(Census.2010, by = join_by(State == State))),
    (grouped_year %>%
    filter(Year == 2019) %>%
    inner_join(Census.2019, by = join_by(State == State)))) %>%
    select(-Average.House.Prices) %>%
    pivot_wider(names_from = Year, values_from = Population) %>%
    dplyr::rename("Population.2010" = "2010", "Population.2019" = "2019")
b3 <-
  rbind(
    (grouped_year %>%
    filter(Year == 2010) %>%
    inner_join(Census.2010, by = join_by(State == State))),
    (grouped_year %>%
    filter(Year == 2019) %>%
    inner_join(Census.2019, by = join_by(State == State)))) %>%
    select(-Population) %>%
    pivot_wider(names_from = Year, values_from = Average.House.Prices) %>%
    dplyr::rename("Average.House.Price.2010" = "2010", "Average.House.Price.2019" = "2019")
Values.2010.2019 <-
  a3 %>% 
  inner_join(b3, by = join_by(State)) %>% 
  select(-Code.y, -Region.y) %>% 
  dplyr::rename("Code" = "Code.x", "Region" = "Region.x") %>%
  filter(State != "District of Columbia")
Values.2010.2019

Values.2010.2019 <- 
  Values.2010.2019 %>%
  arrange(Average.House.Price.2019 - Average.House.Price.2010)

bp11 <- ggplot(data = head(Values.2010.2019, 5), aes(x = State, y = ((Average.House.Price.2019 - Average.House.Price.2010)/Average.House.Price.2010)*100)) +
  geom_bar(stat = "identity", fill = "darkblue", alpha = 0.7) +
  geom_label_repel(aes(label = round(((Average.House.Price.2019 - Average.House.Price.2010)/Average.House.Price.2010)*100, 2)), nudge_x = -0.1, size = 3) +
  theme_classic() +
  ggtitle("Least Price Change") +
  xlab("State") + ylab("Price Difference %") +
  labs(subtitle = "States with the lowest percentage price change between 2010 and 2019")
bp12 <- ggplot(data = tail(Values.2010.2019, 5), aes(x = State, y = ((Average.House.Price.2019 - Average.House.Price.2010)/Average.House.Price.2010)*100)) +
  geom_bar(stat = "identity", fill = "darkgreen", alpha = 0.7) +
  geom_label_repel(aes(label = round(((Average.House.Price.2019 - Average.House.Price.2010)/Average.House.Price.2010)*100, 2)), nudge_x = 0.1, size = 3) +
  theme_classic() +
  ggtitle("Most Price Change") +
  xlab("State") + ylab("Price Difference %") +
  labs(subtitle = "States with the highest percentage price change between 2010 and 2019")
grid.arrange(bp11, bp12, ncol = 2)

Values.2010.2019 <- 
  Values.2010.2019 %>%
  arrange(Population.2019 - Population.2010)

bp13 <- ggplot(data = head(Values.2010.2019, 5), aes(x = State, y = ((Population.2019 - Population.2010)/Population.2010)*100)) +
  geom_bar(stat = "identity", fill = "darkblue", alpha = 0.7) +
  geom_label_repel(aes(label = round(((Population.2019 - Population.2010)/Population.2010)*100, 2)), nudge_x = -0.1, size = 3) +
  theme_classic() +
  ggtitle("Least Population Change") +
  xlab("State") + ylab("Price Difference %") +
  labs(subtitle = "States with the highest percentage population change between 2010 and 2019")
bp14 <- ggplot(data = tail(Values.2010.2019, 5), aes(x = State, y = ((Population.2019 - Population.2010)/Population.2010)*100)) +
  geom_bar(stat = "identity", fill = "darkgreen", alpha = 0.7) +
  geom_label_repel(aes(label = round(((Population.2019 - Population.2010)/Population.2010)*100, 2)), nudge_x = 0.1, size = 3) +
  theme_classic() +
  ggtitle("Most Population Change") +
  xlab("State") + ylab("Price Difference %") +
  labs(subtitle = "States with the highest percentage population change between 2010 and 2019")
grid.arrange(bp13, bp14, ncol = 2)

ggplot(data = Values.2010.2019, aes(x = (Population.2019 - Population.2010), y = (Average.House.Price.2019 - Average.House.Price.2010), color = State)) +
  geom_smooth(method = 'lm', formula = y~x, color = 'darkgrey', alpha = 0.4, linetype = 'dashed', linewidth = 1, se = FALSE) +
  geom_point(alpha = 0.5) + 
  scale_size_continuous(guide = "none") +
  scale_color_discrete(guide = "none") +
  geom_label_repel(aes(label = Code), size = 2, max.overlaps = 20) +
  theme_classic() +
  ggtitle("(2010 and 2019) Trend Between Population and House Prices") +
  xlab("Difference in Population") + ylab("Difference in House Prices") +
  labs(subtitle = "Scatterplot explaining the relationship between the difference in population and the difference in house prices between 2010 and 2019")

ggplot(data = Values.2010.2019, aes(x = (Population.2019 - Population.2010), y = (Average.House.Price.2019 - Average.House.Price.2010), color = State)) +
  geom_smooth(method = 'lm', formula = y~x, color = 'darkgrey', alpha = 0.4, linetype = 'dashed', linewidth = 1, se = FALSE) +
  geom_point(alpha = 0.5) + 
  scale_size_continuous(guide = "none") +
  scale_color_discrete(guide = "none") +
  geom_label_repel(aes(label = Code), size = 2, max.overlaps = 20) +
  facet_wrap(~Region, nrow = 2) +
  theme_classic() +
  ggtitle("(2010 and 2019) Trend Between Population and House Prices") +
  xlab("Difference in Population") + ylab("Difference in House Prices") +
  labs(subtitle = "Scatterplot explaining the relationship between the difference in population and the difference in house prices between 2010 and 2019 faceted by region") +
  theme(strip.text = element_text(size = 8), panel.spacing = unit(1, "lines"))

ggplot(data = Values.2010.2019, aes(x = log((Population.2019 / Population.2010)), y = log((Average.House.Price.2019 / Average.House.Price.2010)), color = State)) +
  geom_hline(yintercept = 0, color = 'black', linetype = 'dotted', linewidth = 0.4) +
  geom_vline(xintercept = 0, color = 'black', linetype = 'dotted', linewidth = 0.4) +
  geom_smooth(method = 'lm', formula = y~x, color = 'darkgrey', alpha = 0.4, linetype = 'dashed', linewidth = 1, se = FALSE) +
  geom_point(alpha = 0.5) + 
  scale_size_continuous(guide = "none") +
  scale_color_discrete(guide = "none") +
  geom_label_repel(aes(label = Code), size = 2, max.overlaps = 30) +
  theme_classic() +
  ggtitle("(2010 and 2019) Trend Between Population and House Prices") +
  xlab("Proportion of Population (log scale)") + ylab("Proportion of House Prices (log scale)") +
  labs(subtitle = "Scatterplot explaining the relationship between the ratio of population and the ratio of house prices between 2010 and 2019")

ggplot(data = Values.2010.2019, aes(x = log((Population.2019 / Population.2010)), y = log((Average.House.Price.2019 / Average.House.Price.2010)), color = State)) +
  geom_hline(yintercept = 0, color = 'black', linetype = 'dotted', linewidth = 0.4) +
  geom_vline(xintercept = 0, color = 'black', linetype = 'dotted', linewidth = 0.4) +
  geom_smooth(method = 'lm', formula = y~x, color = 'darkgrey', alpha = 0.4, linetype = 'dashed', linewidth = 1, se = FALSE) +
  geom_point(alpha = 0.5) + 
  scale_size_continuous(guide = "none") +
  scale_color_discrete(guide = "none") +
  geom_label_repel(aes(label = Code), size = 2, max.overlaps = 30) +
  facet_wrap(~Region, nrow = 2) +
  theme_classic() +
  ggtitle("(2010 and 2019) Trend Between Population and House Prices") +
  xlab("Proportion of Population (log scale)") + ylab("Proportion of House Prices (log scale)") +
  labs(subtitle = "Scatterplot explaining the relationship between the ratio of population and the ratio of house prices between 2010 and 2019 faceted by region") +
  theme(strip.text = element_text(size = 8), panel.spacing = unit(1, "lines"))